Politica sull'Intelligenza Artificiale

v1.0 · 7 maggio 2026

Governance, limitazioni e usi vietati per le funzionalità IA.

Current Status: AI-powered features are planned for Optiviera V2.0 and later releases. This policy establishes the governance framework that will govern AI functionality when introduced. The current platform does not include active AI-assisted features.

Core AI Governance Principles

Human Oversight

AI features operate under human supervision. Automated decisions affecting customers require human review.

Transparency

AI features are clearly identified. Customers are informed when AI processing is involved.

Privacy-Aware

AI processing follows data minimization principles and GDPR-oriented privacy safeguards.

Responsible Use

AI is used only for legitimate, lawful, and authorized purposes. Harmful uses are prohibited.

Accuracy Controls

Output quality monitoring and customer validation workflows to minimize AI errors.

Risk Minimization

AI risks assessed before deployment. High-risk AI applications subject to additional review.

AI Limitations & Accuracy Warning

Important: AI systems are probabilistic technologies. AI-generated outputs may contain inaccuracies, hallucinations, incomplete recommendations, or biased results. Customers must review and validate AI-generated outputs before operational use.

AI outputs are not legal, medical, or financial advice
AI outputs should not be used for regulated automated decision-making
Optiviera does not guarantee factual accuracy of AI outputs
AI features are not suitable for safety-critical applications

AI Data Processing Principles

Data minimization — only process data necessary for AI functionality
Privacy-aware AI integration respecting GDPR principles
Customer data not used to train third-party AI models without explicit consent
AI subprocessors disclosed at /legal/subprocessors
Human oversight maintained for AI-assisted workflows

Prohibited Uses of AI Features

Generating deceptive, misleading, or harmful content
Discriminatory or privacy-violating AI outputs
Circumventing platform security controls
Processing special category data without appropriate legal basis
Using AI outputs as basis for significant decisions without human review

Torna al Centro Fiducia